Relative Pose Information-based 3-D Flocking Control

Abstract

This paper proposes a flocking algorithm on the Special Euclidean group SE(3) for a network of rigid bodies. The main feature of the present law is to embody the three fundamental flocking rules, cohesion, alignment, and separation, presented by Reynolds. We first define rigid body networks and tackle a pose synchronization problem as a partial flocking problem. Then, based on the Reynolds rules, we define flocking motion as the goal of this paper, propose a flocking algorithm and give the convergence analysis. Here, the present scheme is based only on relative pose information to neighbors for imitating actual flocking motion or facility of implementation. We finally show the effectiveness of the present algorithm through simulation.